[WIP] Add support for ESP32S3

devices/esp32/testHTTP.ino

@@ -0,0 +1,80 @@
+#include <Arduino.h>
+#include <WiFi.h>
+#include <HTTPClient.h>
+#include <WiFiClientSecure.h>
+
+const char* ssid = ">ENTER WIFI NAME>"; // Your WiFi SSID
+const char* password = "<ENTER PASS>"; // Your WiFi password
+
+
+// Function to send WAV buffer
+void sendWavBuffer(const std::vector<uint8_t>& buffer) {
+    // Ensure we are connected to WiFi. If not, return.
+    if (WiFi.status() != WL_CONNECTED) {
+        Serial.println("WiFi is not connected!");
+        return;
+    }
+
+    const char* supabaseUrlEnv = "<SUPABASE URL>"; // Use your Supabase URL here
+    const char* authToken = "<ANON TOKEN>"; // Use your Auth Token here
+
+    std::string url = std::string(supabaseUrlEnv) + "/functions/v1/process-audio";
+
+    WiFiClientSecure client;
+    client.setInsecure(); // Disables SSL certificate verification
+
+    HTTPClient https;
+    if (https.begin(client, url.c_str())) { // HTTPS begin
+        https.addHeader("Authorization", "Bearer " + String(authToken));
+        https.addHeader("Content-Type", "audio/wav");
+
+        // Create a non-const copy of the data from the buffer
+        size_t bufferSize = buffer.size();
+        uint8_t* nonConstBuffer = new uint8_t[bufferSize];
+        memcpy(nonConstBuffer, buffer.data(), bufferSize);
+
+        // Use the non-const copy in the POST call
+        int httpResponseCode = https.POST(nonConstBuffer, bufferSize);
+
+        if (httpResponseCode > 0) {
+            // If server responds, print response
+            Serial.printf("HTTP Response code: %d\n", httpResponseCode);
+            String payload = https.getString();
+            Serial.println(payload);
+        }
+        else {
+            Serial.printf("Error on sending POST: %d\n", httpResponseCode);
+        }
+
+        // Clean up the non-const buffer copy after use
+        delete[] nonConstBuffer;
+
+        https.end(); // HTTPS end
+    }
+    else {
+        Serial.println("Error in HTTPS connection");
+    }
+
+
+}
+
+void setup() {
+    Serial.begin(115200);
+
+    // Connect to WiFi
+    WiFi.begin(ssid, password);
+    while (WiFi.status() != WL_CONNECTED) {
+        delay(1000);
+        Serial.println("Connecting to WiFi..");
+    }
+    Serial.println("Connected to WiFi");
+
+    // Initialize vector with dummy data for testing
+    std::vector<uint8_t> dummyData(1024, 0); // Example buffer, replace with actual audio data
+    sendWavBuffer(dummyData); // Send the buffer
+}
+
+void loop() {
+    // Nothing here
+}
+

devices/esp32/testI2S.ino

@@ -0,0 +1,27 @@
+#include "I2S.h"
+
+void setup() {
+  // Open serial communications and wait for port to open:
+  // A baud rate of 115200 is used instead of 9600 for a faster data rate
+  // on non-native USB ports
+  Serial.begin(115200);
+  while (!Serial) {
+    ; // wait for serial port to connect. Needed for native USB port only
+  }
+
+  // start I2S at 16 kHz with 16-bits per sample
+  I2S.setAllPins(-1, 42, 41, -1, -1);
+  if (!I2S.begin(PDM_MONO_MODE, 16000, 16)) {
+    Serial.println("Failed to initialize I2S!");
+    while (1); // do nothing
+  }
+}
+
+void loop() {
+  // read a sample
+  int sample = I2S.read();
+
+  if (sample && sample != -1 && sample != 1) {
+    Serial.println(sample);
+  }
+}

supabase/functions/process-audio/index.ts

@@ -39,9 +39,17 @@ const processAudio = async (req: Request) => {
   try {
     const filenameTimestamp = `adeus_wav_${Date.now()}.wav`;
     const wavFile = await toFile(arrayBuffer, filenameTimestamp);
+
+    const { data, error } = await supabase.storage
+      .from("test")
+      .upload(filenameTimestamp, wavFile);
+
+    if (error) {
+      console.error("Error uploading file:", error);
+    }
 
     const transcriptResponse = await openaiClient.audio.transcriptions.create({
-      file: await toFile(wavFile, filenameTimestamp),
+      file: wavFile,
       model: "whisper-1",
       prompt:
         'Listen to the entire audio file, if no audio is detected then respond with "None" ', // These types of prompts dont work well with Whisper -- https://platform.openai.com/docs/guides/speech-to-text/prompting